- Email: [email protected]
Cognitive processing efficiency in schizophrenia: generalized vs domain specific deficits
SCHIZOPHRENIA RESEARCH ELSEVIER
Schizophrenia Research 30 (1998) 41 4 9
Cognitive processing efficiency in schizophrenia: generalized vs domain specific deficits Jeffrey S c h a t z * Department of Psychology, Washington Universityof St. Louis, St. Louis, Missouri, USA Received 15 February 1997; accepted 21 August 1997
Abstract
The issue of generalized vs specific cognitive deficits in schizophrenia was explored by examining reaction time data from 40 published studies with 196 reaction time conditions. Using a regression-based approach, the proportional relationship between the response times of groups with schizophrenia was compared with those of age-matched, healthy comparison groups. Through this method, the extent to which deficits in processing efficiency are explained by a single factor, general processing speed, was compared with possible domain specific or task specific deficits. The results suggest that, overall, the data conforms well to a general linear slowing model which accounts for 87% of the variance in reaction time performance. Some additional variance, however, is accounted for by different degrees of linear slowing for three types of tasks: tasks involving selective attention/inhibition showed the most slowing (2.3-times slower for schizophrenia), followed by lexical tasks (1.8-times slower for schizophrenia), and finally, non-lexical tasks showed the least slowing (1.4-times slower for schizophrenia). This pattern is distinct from other groups showing generalized slowing, such as older adults, and suggests a unique pattern of information processing deficits in schizophrenia. © 1998 Elsevier Science B.V.
Keywords. Schizophrenia; Processing speed; Reaction time; Inhibition
I. Introduction
The extent to which disruption of cognition in schizophrenia is better characterized by global vs specific deficits has been difficult to establish. Though deficits across m a n y types of traditional cognitive measures are evident, specific deficits in schizophrenia are frequently noted in attention, m e m o r y and executive skills (for review, see * Present address: Department of Psychiatry, University of California, San Francisco, 401 Parnassus Ave, Box CPT, San Francisco, CA 94143, USA. Tel. 415 476 7691; Fax: 415 476 7719; e-mail: [email protected] 0920-9964/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S0920-9964 ( 9 7 ) 0 0 1 2 5 - 4
Randolph et al., 1993). Reaction time paradigms from experimental psychology have also frequently been used with the disorder to study specific cognitive processes. As with more traditional cognitive measures, some authors have proposed that the cognitive deficits seen in schizophrenia may not be specific to any particular cognitive process, but are more general in nature (e.g. Gjerde, 1983; Yates, 1966). As an alternative, these researchers suggest there may be a general slowness on information processing tasks which leads to increasingly slowed performance as the complexity of cognitive tasks increases. Most studies using reaction time paradigms, however, typically look for specific cogni-
42
J. Schatz / Schizophrenia Research 30 (1998) 41-49
tive deficiencies, and therefore, examine only a limited number of types of information processing (e.g., tasks primarily of attentional functions). Because these studies usually test a limited range of cognitive skills, the extent of generalized slowing of cognitive processes is not well tested. A non-traditional method for evaluating the generalized slowing of cognition exists that has been well developed in the adult aging literature. Using a regression based approach, the response time of an older adult group is expressed as a function of the response time of a younger adult group across a range of reaction time tasks (see Cerella, 1985). A regression line can be plotted from this function in order to express the relationship between the two groups' data. If the two groups' data were equivalent, the regression line would have a slope of 1 and an intercept of zero. With this approach, sensory or motor components that are common to all of the tasks add a constant degree of slowing across all tasks. Such common factors influence the intercept of the regression line. The slope of the regression line, however, represents proportional increases in processing time that are dependent on the cognitive complexity of the task. Thus, the slope is considered a measure of general cognitive processing efficiency. Finally, individual response time conditions that lie significantly far from the regression line represent specific tasks which are slowed more (or less) than would be predicted from general cognitive slowing. Reviews of reaction time studies have demonstrated that much of the difference in cognitive performance between older and younger adults can be expressed in terms of a single, linear relationship between the two groups' response times. This single factor, generalized slowing, accounts for 89-98% of the variability in performance across tasks (Cerella, 1985). That is, as the complexity of a cognitive task increases, the relationship between the performance of young adults and older adults differs by the same proportional amount, regardless of the specific task content (e.g., older adults are 1.5-times slower across tasks). One significance of this finding is that, for many cognitive abilities, what were previously believed to represent agerelated deficits in specific cognitive operations were
most likely part of a general deficit in processing speed (for examples of such attributional errors, see Cerella, 1985; and Salthouse, 1985). A second contribution of this method is that it allows for one to separately evaluate sensory-motor vs cognitive aspects of slowed response speed. The specific contribution of each of these factors to psychomotor slowing is less clear in individual cognitive tasks. The purpose of the following report is to explore the extent to which cognitive processing in schizophrenia can be represented by generalized slowing. As has been used previously in exploring this concept in adult aging, child development, and Alzheimer's disease, a meta-analytic approach is used (see Cerella, 1985; Kail, 1991; Nebes and Madden, 1988). This method appears to be useful because the results of meta-analyses have generally conformed to the results of experimental studies (Cerella, 1985; Hale, 1990; Hale and Myerson, 1996; Kail, 1991; Lima et al., 1991). In addition, recent reports have indicated processing speed may not be a unitary construct. In adult aging, a distinction between lexical/verbal and non-lexical/non-verbal processing tasks has been shown to be important, as older adults show greater slowing on tasks requiring non-lexical information processing (e.g. mental rotation, visual search) as compared with lexical processing tasks (e.g. lexical decision: see Lima et al., 1991; Hale and Myerson, 1996). Because of this potentially useful distinction, differences in the extent of slowing for selected domains of processing were also explored to test if cognitive speed may be disproportionately affected for particular types of processing.
2. Data analysis The studies included in the meta-analysis were located using Medline and Psychlit data bases for the period 1974 1995. The key word 'schizophrenia' was combined separately with the key words 'reaction time', 'lexical', 'semantic', and 'inhibition'. Studies were located which included: (a) both a schizophrenia group and a healthy, agematched, non-psychiatric comparison group; (b)
J. Schatz / Schizophrenia Research 30 (1998) 41 49
a cognitive reaction time task requiring a manual or vocal response to frequently occurring stimuli; and (c) mean reaction time values for each group. Forty studies were located which included 196 different conditions (see Appendix A). The types of cognitive tasks in these studies varied considerably, however, they can generally be described as tasks of simple and choice reaction time, letter or digit identification, letter matching, single word naming, lexical decision, the Stroop paradigm, size judgments, visual search, picture matching, judgment of facial expression and memory scanning. The mean age of the groups with schizophrenia was 30.9 years and for the age-matched comparison participants 29.9 years (reported in 40 and 38 studies, respectively), the mean of mean years of education for the schizophrenia groups was 12.3 and for the comparison groups 12.8 (reported for 18 and 16 studies, respectively), and the mean of the mean percent of males was 80% for schizophrenia groups and 74% for the comparison groups (reported for 31 and 32 studies, respectively). Lexical tasks were defined as conditions requiring the retrieval of names or words from either short-term or long-term memory. The majority of the lexical tasks were lexical decision paradigms. Non-lexical tasks involved the processing of non-language based material such as geometric shapes or visual search paradigms. A third set of tasks which did not fit well into the lexical/non-lexical domains was also apparent in this set of studies. These tasks all required the suppression of more salient, competing stimuli in order to process the target stimuli. These tasks all appeared to involve components of selective attention and/or inhibition, such as in the incongruent condition of the Stroop paradigm. In the final coding of tasks, therefore, three domains were used: tasks of lexical processing, non-lexical processing and selective attention/inhibition. All conditions were classified separately by the author and a second coder. The two coders agreed on their classification of 188 of 196 conditions. Discrepancies were resolved by discussion between the two coders. Mean reaction time values were first weighted according to the number of subjects in the group (weighted by the square root of n - 1 ) in order to
43
give greater weighting to more reliable reaction time values. It was later determined that analyzing the data with or without weighting, however, did not change the pattern of results or the processing speed equations in any meaningful way. The data are therefore reported without weightings. To examine the extent to which cognitive performance could be described by generalized slowing, a hierarchical multiple regression procedure was used, with the mean reaction time performance of the schizophrenia groups as the dependent variable. In the first step, the mean reaction time performance of the comparison groups was entered to determine the linear relationship between the schizophrenia and comparison group response times, R2=0.87, F(1,194)=1341.33, p<0.01. In the second step, the distinction between types of tasks (lexical, non-lexical, and inhibitory) was entered with the three types of tasks dummy coded, I R 2 = 0 . 0 I , F(2,192)--4.57, p<0.05. This second step statistically adjusts for any differences in overall response time across conditions, irrespective of task complexity. In the third step, the interaction terms were entered to determine if the regression lines differed between lexical, non-lexical and inhibitory tasks, IRZ=0.02, F(2,190)= 20.36, p<0.01. An examination of Cooke's distance and standardized residuals from the final regression model indicated no condition had a Cooke's distance of greater than 0.24 or a standardized residual greater than 2.4, suggesting there were no significant outliers present. Because of the presence of a significant interaction, three linear regression procedures were conducted as post hoc analyses to examine the nature of the interactions. The results of these procedures is shown in Table 1. The significant interaction terms for each comparison indicated each task type had a different slope, suggesting separate processing speed factors for lexical, nonlexical, and inhibitory tasks. The line of best fit for lexical tasks was y = 1.75x-240, R z =0.75, for non-lexical tasks y = 1.39x+7, R2=0.93, and for inhibitory tasks y = 2 . 3 0 x - 5 8 5 , R2=0.88 (see Figs. 1-3). It should be noted, however, that overall the data were well fit with a single linear regression model, y = 1.50x - 52, R 2 = 0.87. Uneven distribution of reaction time values has
44
J. Schatz / Schizophrenia Research 30 (1998) 41 49
Table 1 Hierarchical regression procedures comparing the regression lines for lexical, nonlexical, and inhibitory processing speed equations for schizophrenia Independent variable
df
R2
IR 2
F
A. Lexical vs non-lexical step 1: RT-comparison group step 2: task type step 3: RT x task type
l 177 1 776 1 175
0.88 0.88 0.89
0.88 0.00 0.01
1340.66"* 0.54 12.90"*
B. Lexical vs inhibitory step 1: RT-comparison group step 2: task type step 3: RT x task type
1 102 1 101 I 100
0.80 0.80 0.82
0.80 0.01 0.01
402.36** 2.78 10.11"*
C. Nonlexical vs inhibitory step 1: RT-comparison group step 2: task type step 3: RT x task type
l 107 1 106 1 105
0.91 0.91 0.94
0.91 0.00 0.03
1022.19"* 5.74* 40.89**
*p < 0.05.
**p<0.01. 2500
25007
[I i
2o00J
%
2000-
~ ,5oo!
Isoo~ z /
.~
.~ ,ooo]
I000~
500
~ [] D ,
© 0
!
5OO
j
/ ,/ /,
0
/
0 500
1000
1500
2000
2500
RT (mse¢) - comparison group
0
500
1000
1500
2000
2500
RT (mscc) - comparison group
Fig. 1. Scatterplot of reaction time values showing the lines of best fit for non-lexical cognitive tasks. The linear model of 1.39x+7 accounts for 93% of the variance. The slope differs significantly from 1, t(88)= 12.29, p<0.01, indicating slowing on non-lexical tasks relative to the comparison group.
Fig. 2. Scatterplot of reaction time values showing the lines of best fit for lexical cognitive tasks. The linear model of 1.75x-240 accounts for 75% of the variance. The slope differs significantly from 1, t(85)=6.77, p<0.01, indicating slowing on lexical tasks relative to the comparison group.
b e e n r a i s e d as a p o t e n t i a l c o n f o u n d w h e n c o m p a r ing d i f f e r e n t d o m a i n s o f r e a c t i o n t i m e t a s k s in general slowing models. Because the reaction time values for non-lexical tasks had a number of s h o r t e r r e a c t i o n t i m e v a l u e s t h a n e i t h e r t h e lexical o r i n h i b i t o r y t a s k s (i.e., r e a c t i o n t i m e v a l u e s in the c o m p a r i s o n g r o u p o f less t h a n 350 m s ) , the regres-
s i o n a n a l y s e s w e r e r e p e a t e d after d r o p p i n g t h e s e exceptionally short reaction time conditions from the non-lexical domain. This created equal ranges o f r e a c t i o n t i m e v a l u e s in the c o m p a r i s o n g r o u p s a c r o s s all t h r e e d o m a i n s . T h i s m a t c h i n g p r o c e d u r e , however, had a minimal impact on the hierarchical r e g r e s s i o n p r o c e d u r e s : the i n h i b i t o r y d o m a i n still
J. Schat: / Schizophrenia Research 30 (1998) 41-49 2500
2000-
,2.
©
1000 ~
/
g
/ A /a
, 500
/
/
,/ 500
1000
1500
2000
2500
RT (msec) - comparison group
Fig. 3. Scatterplot of reaction time values showing the lines of best fit for inhibitory cognitive tasks. The linear model of 2.30x-585 accounts for 88% of the variance. The slope differs significantly from 1, t(15)=6.06, p<0.01, indicating slowing on inhibitory tasks relative to the comparison group. differed from the lexical and non-lexical domains; the difference between the lexical and non-lexical domains, however, only reached a trend level ( p < 0 . 0 6 ) . This matching procedure did not significantly alter the processing speed equation for the non-lexical domain (y = 1.36x + 18). A second set of analyses was conducted to explore possible relationships between generalized slowing and particular demographic or clinical aspects of the groups with schizophrenia. From the 198 conditions, the following variables were coded for each group: age of the schizophrenia group (available for 198 conditions); years of education in the schizophrenia group (available for 135 conditions); percent of males in the sample (available for 168 conditions); mean total score on the Brief Psychiatric Rating Scale for the sample (BPRS, available for 77 conditions); the percent of individuals classified as having paranoid features (available for 42 conditions); and the percent of the sample on antipsychotic medication at the time of the study (available for 145 conditions). Standardized residuals for each condition were saved from the linear regression model exploring generalized slowing (i.e., the first regression analysis reported above). These values represent the extent to which each reaction time value was
45
slower (or faster) than predicted from the overall generalized slowing model. Correlation coefficients between these standardized residuals and the demographic and clinical variables indicated relatively slower response times in the schizophrenia group compared with the comparison group were associated with lower education, r = - 0.47, p < 0 . 0 1 , higher BPRS scores, r--0.48, p<0.01, a lower percent of paranoid-type individuals, r = - 0.52, p < 0.01, and a higher percent of individuals on antipsychotic medications, r=0.30, p <0.01. Finally, an analysis was conducted of the abovementioned demographic and clinical factors in relation to the cognitive domains sampled in the study. This analysis evaluated if the domain specific patterns of cognitive slowing could be due to an interaction between characteristics of the patient population and the type of task used in the study. Each demographic or clinical variable was used as the dependent variable and the cognitive domain for each condition (lexical, non-lexical, inhibitory) was treated as the independent variable, One-way analysis of variance procedures did not indicate demographic characteristics were related to the type of task used in the studies (all Fs less than 1).
3.
Discussion
The current meta-analysis suggests several points regarding cognitive processing efficiency in schizophrenia. First, generalized slowing appears to be a significant aspect of information processing in schizophrenia, accounting for 87% of the variance in reaction times across studies. Thus, in terms of effect size, generalized slowing in schizophrenia is a highly significant aspect of performance. This general slowing of cognitive processing speed may have important implications for understanding the sources of cognitive dysfunction in schizophrenia. General processing speed has been shown to be a primary mechanism for improvements in working memory capacity in children (Fry and Hale, 1996; Kail, 1992; Kail and Park, 1994). The importance of processing speed for the development of working memory suggests that
46
J. Schatz / Schizophrenia Research 30 (1998) 41-49
understanding the mechanisms responsible for cognitive slowing in schizophrenia could provide a developmental framework for working memory dysfunction in this disorder. The presence of working memory dysfunction in both individuals with schizophrenia and first-degree relatives (Park et al., 1995; Spitzer, 1993) further suggests that abnormalities in processing speed could be a part of the genotype of the disorder. Further study is needed to evaluate these hypotheses. A second implication of this meta-analysis is that, although a single factor accounts for a large proportion of the variance, the data are better fit by also considering three domains of cognition. Thus, the extent of cognitive slowing appears to be dependent, in part, on the particular type of information processing. The domain specific aspects of slowing may be of theoretical and clinical importance, but appear to be small in terms of effect size. There were no significant outlier conditions present in this meta-analysis, which suggests that although domains may differ in degree of slowing, there was no evidence for a specific cognitive task which showed disproportionate slowing beyond domain specific effects. Certainly, not all potential cognitive domains or specific cognitive processes were represented in the meta-analysis. Nonetheless, the data suggest that not all classes of cognitive processes are impacted evenly in schizophrenia. This is counter to a hypothesis that the cognitive dysfunction in schizophrenia is due to a single, generalized deficit. A third implication of the results of this study is that the pattern of slowing among these three domains has not been observed previously in groups with significant cognitive slowing, such as older adults (Hale et al., 1987; Lima et al., 1991). This suggests the pattern of cognitive slowing in schizophrenia may be a unique configuration of inefficiencies that differs from other types of cognitive slowing. The particularly inefficient processing on tasks which required inhibition, or ignoring competing stimuli, is consistent with previous suggestions that the breakdown of normal selective attention mechanisms is an important aspect of cognition in schizophrenia (Shakow, 1963). There are several points to be made regarding the phenomenon of generalized slowing in schizo-
phrenia. First, as has been done in adult aging, it will be useful to experimentally validate the present pattern of findings. Second, although cerebral dysfunction is associated with cognitive slowing (Ferraro, 1996), little is known about the relationship between domain specific slowing and cerebral dysfunction. Future studies using functional imaging or examining individuals with focal injury may be useful for understanding the neurologic underpinnings of domain specific slowing in schizophrenia. Learning more about these systems could help further illuminate some specific types of cerebral dysfunction in schizophrenia. Finally, because schizophrenia is a heterogeneous disorder, it may be useful to explore variability in the pattern of cognitive slowing across domains in relation to other factors, such as symptomatology or variations in the course of the disorder. An exploratory analysis of such factors in the present meta-analysis suggested there may be aspects of the severity and symptomatology of schizophrenia associated with the extent of generalized slowing. The inconsistent reporting of these variables in the studies reviewed in this report, however, makes it difficult to draw strong conclusions about the precise nature of these relationships. This meta-analysis also suggests several methodological considerations when using reaction time to measure cognitive performance in schizophrenia. As has been discussed with accuracy based measures of cognition (e.g., Chapman and Chapman, 1989), it is extremely important to consider task difficulty, in addition to task content. If only a limited number of reaction time measures are to be used, specific cognitive abilities will be best understood if one matches different cognitive tasks for response time in a healthy comparison group. A second strategy to control for cognitive slowing is to identify cognitive processes whereby poorer performance is associated with faster response times. This strategy for separating out generalized slowing from task specific deficits may prove particularly fruitful when studying normal inhibitory processes, as one may expect faster reaction times with defective inhibition across a number of cognitive paradigms (see Dagenbach and Carr, 1994). It is also recommended that researchers expand the range of cognitive reaction
J. Schatz / Schizophrenia Research 30 (1998) 41 49
time tasks used in individual studies. This would strengthen conclusions in individual studies that any task specific effects found are truly above and beyond what is expected from generalized slowing.
Appendix A Studies included in meta-analysis
Bruder, G., Sutton, S., Babkoff, H., Yozawitz, A., Fleiss, J.L., 1975. Auditory signal detectability and facilitation of reaction time in psychiatric patients and nonpatients. Psychological Medicine, 5, 260-272. Bruder, G., Yozawitz, A., Berenhaus, I., Sutton, S., 1980. Reaction time facilitation in affective psychotic patients. Psychological Medicine, 10, 549-554. Burch, J.W., 1995. Typicality range deficit in schizophrenics' recognition of emotion in faces. Journal of Clinical Psychology, 51, 140-52. Carter, C.S., Robertson, L.C., Nordahl, T.E., 1992. Abnormal processing of irrelevant information in chronic schizophrenia: selective enhancement of Stroop facilitation. Psychiatry Research, 4l, 137-146. Carter, C.S., Robertson, L.C., Nordahl, T.E., O'Shora-Celaya, L.J. and Chaderjian, M.C., 1993. Abnormal processing of irrelevant information in schizophrenia: the role of illness subtype. Psychiatry Research, 48, 17-26. Chapin, K., Wightma, L., Lycaki, H., Josef, N., Rosenbaum, G., 1987. American Journal of Psychiatry, 144, 948-950. David, A.S., 1995. Negative priming (cognitive inhibition) in psychiatric patients: effects of neuroleptics. Journal of Nervous and Mental Disease, 183, 337-339. David, A.S., 1993. Spatial and selective attention in the cerebral hemispheres in depression, mania and schizophrenia. Brain and Cognition, 23, 166-180. Davies-Osterkamp, S., Rist, F., Bangert, A., 1977. Selective attention, breadth of attention, and shifting attention in chronic nonparanoid schizophrenics. Journal of Abnormal Psychology, 86, 461-469.
47
Elkins, I.J., Cromwell, R.L., 1994. Priming effects in schizophrenia: associative interference and facilitation as a function of visual context. Journal of Abnormal Psychology, 103, 791-800. Gray, A. L., 1975. Autonomic correlates of chronic schizophrenia: a reaction time paradigm. Journal of Abnormal Psychology, 84, 189-196. Hermanutz, M., Gestrich, J (1991. Computerassisted attention training in schizophrenia. European Archives of Psychiatry and Clinical Neuroscience, 240, 282-287. Hirt, M., Pithers, W., 1990. Arousal and maintenance of schizophrenic attention. Journal of Clinical Psychology, 46, 15-20. Hirt, M., Pithers, W., 1991. Selective attention and levels of coding in schizophrenia. British Journal of Clinical Psychology, 30, 139-149. John, C.H., Hemsley, D.R., 1992. Gestalt perception in schizophrenia. European Archives of Psychiatry and Clinical Neuroscience, 241, 215-221. Knight, R.G., Youard, P.J., Wooles, I.M., 1985. Visual information-processing deficits in chronic schizophrenic subjects using tasks matched for discriminating power. Journal of Abnormal Psychology, 94, 454-459. Koh, S.D., Szoc, R., Peterson, R.A., 1977. Shortterm memory scanning in schizophrenic young adults. Journal of Abnormal Psychology, 86, 451-460. Kopp, B., Mattler, U., Rist, F., 1994. Selective attention and response competition in schizophrenia patients. Psychiatry Research, 53, 129-139. Koyama, S., Hokama, H., Miyatani, M., Ogura, C., Nageishi, Y., Shimokochi, M., 1994. Electroencephalography and Clinical Neurophysiology, 92, 546-554. Koyama, S., Nageishi, Y., Minoru, S., Hokama, H., Yoshikazu, M., Miyatani, M., Ogura, C., 1991. The N400 component of event-related potentials in schizophrenia patients: a preliminary study. Electroencephalography and Clinical Neurophysiology, 78, 124-132. Laplante, L., Everett, J., Thomas, J., 1992. Inhibition through negative priming with Stroop stimuli in schizophrenia. British Journal of Clinical Psychology, 31,307-326.
48
J. Schatz / Schizophrenia Research 30 (1998) 41-49
Maier, W., Franke, P., Kopp, B., Hardt, J., Hain, Ch., Rist, F., 1994. Reaction time paradigms in subjects at risk for schizophrenia. Schizophrenia Research, 13, 35-43. Min, S.K., Oh, B.H., 1992. Hemispheric asymmetry in visual recognition of words and motor response in schizophrenic and depressive patients. Biological Psychiatry, 31,255-262. Ober, B.A., Vinogradov, S., Shenaut, G.K., 1995. Semantic priming of category relations in schizophrenia. Neuropsychology, 9, 220-228. Pharr, D.R., Connor, J.M., 1977. Similarities and differences in encoding processes in chronic schizophrenics and normals. Perceptual and Motor Skills, 45, 43 t-443. Pharr, D.R., Connor, J.M., 1980. Memory scanning in a visual search task by schizophrenics and normals. Journal of Clinical Psychology, 36, 625-631. Rist, F., Thurm, I., 1984. Effects of intramodal and crossmodal stimulus diversity on the reaction time of chronic schizophrenics. Journal of Abnormal Psychology, 93, 331-338. Rockstroh, B., Muller, M., Wagner, M., Cohen, R., Elbert, T., 1994. Event-related and motor response to probes in a forewarned reaction time task in schizophrenic patients. Schizophrenia Research, 13, 23-34. Rosenbaum, G., Shore, D,, Chapin, K., 1988. Attention deficit in schizophrenia and schizotypy: marker versus symptom variables. Journal of Abnormal Psychology, 97, 41-47. Rosofsky, I., Levin, S., Holzman, P.S., 1982. Psychomotility in the functional psychoses. Journal of Abnormal Psychology, 91, 71-74. Schwartz, F., Carr, A.C., Munich, R.L., Glauber, S., Lesser, B., Murray, J., 1989. Reaction time impairment in schizophrenia and affective illness: the role of attention. Biological Psychiatry, 25, 540-548. Spitzer, M., Braun, U., Maier, S., Hermle, L., Maher, B.A., 1993. Indirect semantic priming in schizophrenic patients. Schizophrenia Research, 12, 71-80. Spitzer, M., Braun, U., Hermle, L., Maier, S., 1993. Associative semantic network dysfunction in thought-disordered schizophrenic patients:
direct evidence from indirect semantic priming. Biological Psychiatry, 34, 864-877. Spitzer, M., Weisker, I., Winter, M., Maier, S., Hermle, L., Maher, B.A., 1994. Semantic and phonological priming in schizophrenia. Journal of Abnormal Psychology, 103, 485-494. Strandburg, R.J., Marsh, J.T., Brown, W.S., Asarnow, R.G., Futhrie, D., Higa, J., 1991. Reduced attention-related negative potentials in schizophrenic children. Electroencephalography and Clinical Neurophysiology, 79, 291 307. Vensky-Stalling, I., Mussgay, L., Cohen, R., 1985. Cognitive dysfunction in chronic schizophrenia on a letter matching task by Posner. Zeitschrift fur Klinische Psychologie, 14, 228 237. Vinogradov, S., Ober, B.A., Shenaut, G.K., 1992. Semantic priming of word pronunciation and lexical decision in schizophrenia. Schizophrenia Research, 8, 171-181. White, C., Farley, J., Charles, P., 1987. Chronic schizophrenic disorder II. Reaction time, social performance and arousal. British Journal of Psychiatry, 150, 374 379. Zahn, T.P., Carpenter, W.T., McGlashan, T.H., 1981. Autonomic nervous system activity in acute schizophrenia: method and comparison with normal controls. Archives of General Psychiatry, 38, 251-258. Zahn, T.P., Rumsey, J.M., Van Kammen, D.P., 1987. Autonomic nervous system activity in autistic, schizophrenic, and normal men: effects of stimulus significance. Journal of Abnormal Psychology, 96, 135-144.
References
Cerella, J., 1985. Information processing rates in the elderly. Psychol. Bull. 98, 67 83. Chapman, L.J., Chapman, J.P., 1989. Strategies for resolving the heterogeneityof schizophrenics and their relatives using cognitive measures. J. Abnorm. Psychol. 98, 357-366. Dagenbach, D., Carr, T.H., 1994. Inhibitory Processes in Attention, Memory, and Language. Academic Press, San Diego. Ferraro, R.F., 1996. Cognitive slowing in closed head injury. Brain and Cognition 32, 429M40. Fry, A.F., Hale, S., 1996. Processingspeed, working memory, and fluid intelligence: evidencefor a developmentalcascade. Psychol. Sci. 7, 237 241.
J. Schatz / Schizophrenia Research 30 (1998) 41-49
Gjerde, P.F., 1983. Attention capacity dysfunction and arousal in schizophrenia. Psychol. Bull. 93, 57-72. Hale, S., 1990. A global developmental trend in cognitive processing speed. Child Development 61,653 663. Hale, S., Myerson, J., 1996. Experimental evidence for differential slowing in the lexical and nonlexical domains. Aging, Neuropsychology, and Cognition 2, 154 165. Hale, S., Myerson, J., Wagstaff, D., 1987. General slowing of nonverbal information processing speed: evidence of a power law. J. Gerontol. 42, 131 136. Kail, R., 1991. Developmental change in speed of processing during childhood and adolescence. Psychol. Bull. 109, 490 501. Kaik R., 1992. Processing speed, speech rate, and memory. Dev. Psychol. 28, 899 904. Kail, R., Park, Y., 1994. Processing time, articulation time, and memory span. J. Exp. Child Psychol. 57, 281 291. Lima, S.D., Hale, S., Myerson, J., 1991. How general is general slowing? Evidence from the lexical domain. Psychol. Aging 6, 416 425.
49
Nebes, R.D., Madden, D.J., 1988. Different patterns of cognitive slowing produced by Alzheimer's disease and normal aging. Psychol. Aging 3, 102-104. Park, S., Holzman, P.S., Goldman-Rakic, P.S., 1995. Spatial working memory deficits in the relatives of schizophrenic patients. Arch. Gen. Psychiatry 52, 821-828. Randolph, C., Goldberg, T.E., Weinberger, D.R., 1993. The neuropsychology of schizophrenia. In: Heilman, K.M., Valenstein, E. (Eds), Clinical Neuropsychology. Oxford University Press, New York, pp. 499-522. Salthouse, T.A., 1985. A Theory of Cognitive Aging. NorthHolland, Amsterdam. Shakow, D., 1963. Psychological deficit in schizophrenia. Behav. Sci. 8, 275-305. Spitzer, M., 1993. The psychopathology, neuropsychology, and neurobiology of associative and working memory deficits in schizophrenia. Eur. Arch. Psychiatry Clin. Neurosci. 243, 57 70. Yates, A.J., 1966. Psychological deficit. Annual Review in Psychology 18, 103 117.
Schizophrenia Research 30 (1998) 41 4 9
Cognitive processing efficiency in schizophrenia: generalized vs domain specific deficits Jeffrey S c h a t z * Department of Psychology, Washington Universityof St. Louis, St. Louis, Missouri, USA Received 15 February 1997; accepted 21 August 1997
Abstract
The issue of generalized vs specific cognitive deficits in schizophrenia was explored by examining reaction time data from 40 published studies with 196 reaction time conditions. Using a regression-based approach, the proportional relationship between the response times of groups with schizophrenia was compared with those of age-matched, healthy comparison groups. Through this method, the extent to which deficits in processing efficiency are explained by a single factor, general processing speed, was compared with possible domain specific or task specific deficits. The results suggest that, overall, the data conforms well to a general linear slowing model which accounts for 87% of the variance in reaction time performance. Some additional variance, however, is accounted for by different degrees of linear slowing for three types of tasks: tasks involving selective attention/inhibition showed the most slowing (2.3-times slower for schizophrenia), followed by lexical tasks (1.8-times slower for schizophrenia), and finally, non-lexical tasks showed the least slowing (1.4-times slower for schizophrenia). This pattern is distinct from other groups showing generalized slowing, such as older adults, and suggests a unique pattern of information processing deficits in schizophrenia. © 1998 Elsevier Science B.V.
Keywords. Schizophrenia; Processing speed; Reaction time; Inhibition
I. Introduction
The extent to which disruption of cognition in schizophrenia is better characterized by global vs specific deficits has been difficult to establish. Though deficits across m a n y types of traditional cognitive measures are evident, specific deficits in schizophrenia are frequently noted in attention, m e m o r y and executive skills (for review, see * Present address: Department of Psychiatry, University of California, San Francisco, 401 Parnassus Ave, Box CPT, San Francisco, CA 94143, USA. Tel. 415 476 7691; Fax: 415 476 7719; e-mail: [email protected] 0920-9964/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S0920-9964 ( 9 7 ) 0 0 1 2 5 - 4
Randolph et al., 1993). Reaction time paradigms from experimental psychology have also frequently been used with the disorder to study specific cognitive processes. As with more traditional cognitive measures, some authors have proposed that the cognitive deficits seen in schizophrenia may not be specific to any particular cognitive process, but are more general in nature (e.g. Gjerde, 1983; Yates, 1966). As an alternative, these researchers suggest there may be a general slowness on information processing tasks which leads to increasingly slowed performance as the complexity of cognitive tasks increases. Most studies using reaction time paradigms, however, typically look for specific cogni-
42
J. Schatz / Schizophrenia Research 30 (1998) 41-49
tive deficiencies, and therefore, examine only a limited number of types of information processing (e.g., tasks primarily of attentional functions). Because these studies usually test a limited range of cognitive skills, the extent of generalized slowing of cognitive processes is not well tested. A non-traditional method for evaluating the generalized slowing of cognition exists that has been well developed in the adult aging literature. Using a regression based approach, the response time of an older adult group is expressed as a function of the response time of a younger adult group across a range of reaction time tasks (see Cerella, 1985). A regression line can be plotted from this function in order to express the relationship between the two groups' data. If the two groups' data were equivalent, the regression line would have a slope of 1 and an intercept of zero. With this approach, sensory or motor components that are common to all of the tasks add a constant degree of slowing across all tasks. Such common factors influence the intercept of the regression line. The slope of the regression line, however, represents proportional increases in processing time that are dependent on the cognitive complexity of the task. Thus, the slope is considered a measure of general cognitive processing efficiency. Finally, individual response time conditions that lie significantly far from the regression line represent specific tasks which are slowed more (or less) than would be predicted from general cognitive slowing. Reviews of reaction time studies have demonstrated that much of the difference in cognitive performance between older and younger adults can be expressed in terms of a single, linear relationship between the two groups' response times. This single factor, generalized slowing, accounts for 89-98% of the variability in performance across tasks (Cerella, 1985). That is, as the complexity of a cognitive task increases, the relationship between the performance of young adults and older adults differs by the same proportional amount, regardless of the specific task content (e.g., older adults are 1.5-times slower across tasks). One significance of this finding is that, for many cognitive abilities, what were previously believed to represent agerelated deficits in specific cognitive operations were
most likely part of a general deficit in processing speed (for examples of such attributional errors, see Cerella, 1985; and Salthouse, 1985). A second contribution of this method is that it allows for one to separately evaluate sensory-motor vs cognitive aspects of slowed response speed. The specific contribution of each of these factors to psychomotor slowing is less clear in individual cognitive tasks. The purpose of the following report is to explore the extent to which cognitive processing in schizophrenia can be represented by generalized slowing. As has been used previously in exploring this concept in adult aging, child development, and Alzheimer's disease, a meta-analytic approach is used (see Cerella, 1985; Kail, 1991; Nebes and Madden, 1988). This method appears to be useful because the results of meta-analyses have generally conformed to the results of experimental studies (Cerella, 1985; Hale, 1990; Hale and Myerson, 1996; Kail, 1991; Lima et al., 1991). In addition, recent reports have indicated processing speed may not be a unitary construct. In adult aging, a distinction between lexical/verbal and non-lexical/non-verbal processing tasks has been shown to be important, as older adults show greater slowing on tasks requiring non-lexical information processing (e.g. mental rotation, visual search) as compared with lexical processing tasks (e.g. lexical decision: see Lima et al., 1991; Hale and Myerson, 1996). Because of this potentially useful distinction, differences in the extent of slowing for selected domains of processing were also explored to test if cognitive speed may be disproportionately affected for particular types of processing.
2. Data analysis The studies included in the meta-analysis were located using Medline and Psychlit data bases for the period 1974 1995. The key word 'schizophrenia' was combined separately with the key words 'reaction time', 'lexical', 'semantic', and 'inhibition'. Studies were located which included: (a) both a schizophrenia group and a healthy, agematched, non-psychiatric comparison group; (b)
J. Schatz / Schizophrenia Research 30 (1998) 41 49
a cognitive reaction time task requiring a manual or vocal response to frequently occurring stimuli; and (c) mean reaction time values for each group. Forty studies were located which included 196 different conditions (see Appendix A). The types of cognitive tasks in these studies varied considerably, however, they can generally be described as tasks of simple and choice reaction time, letter or digit identification, letter matching, single word naming, lexical decision, the Stroop paradigm, size judgments, visual search, picture matching, judgment of facial expression and memory scanning. The mean age of the groups with schizophrenia was 30.9 years and for the age-matched comparison participants 29.9 years (reported in 40 and 38 studies, respectively), the mean of mean years of education for the schizophrenia groups was 12.3 and for the comparison groups 12.8 (reported for 18 and 16 studies, respectively), and the mean of the mean percent of males was 80% for schizophrenia groups and 74% for the comparison groups (reported for 31 and 32 studies, respectively). Lexical tasks were defined as conditions requiring the retrieval of names or words from either short-term or long-term memory. The majority of the lexical tasks were lexical decision paradigms. Non-lexical tasks involved the processing of non-language based material such as geometric shapes or visual search paradigms. A third set of tasks which did not fit well into the lexical/non-lexical domains was also apparent in this set of studies. These tasks all required the suppression of more salient, competing stimuli in order to process the target stimuli. These tasks all appeared to involve components of selective attention and/or inhibition, such as in the incongruent condition of the Stroop paradigm. In the final coding of tasks, therefore, three domains were used: tasks of lexical processing, non-lexical processing and selective attention/inhibition. All conditions were classified separately by the author and a second coder. The two coders agreed on their classification of 188 of 196 conditions. Discrepancies were resolved by discussion between the two coders. Mean reaction time values were first weighted according to the number of subjects in the group (weighted by the square root of n - 1 ) in order to
43
give greater weighting to more reliable reaction time values. It was later determined that analyzing the data with or without weighting, however, did not change the pattern of results or the processing speed equations in any meaningful way. The data are therefore reported without weightings. To examine the extent to which cognitive performance could be described by generalized slowing, a hierarchical multiple regression procedure was used, with the mean reaction time performance of the schizophrenia groups as the dependent variable. In the first step, the mean reaction time performance of the comparison groups was entered to determine the linear relationship between the schizophrenia and comparison group response times, R2=0.87, F(1,194)=1341.33, p<0.01. In the second step, the distinction between types of tasks (lexical, non-lexical, and inhibitory) was entered with the three types of tasks dummy coded, I R 2 = 0 . 0 I , F(2,192)--4.57, p<0.05. This second step statistically adjusts for any differences in overall response time across conditions, irrespective of task complexity. In the third step, the interaction terms were entered to determine if the regression lines differed between lexical, non-lexical and inhibitory tasks, IRZ=0.02, F(2,190)= 20.36, p<0.01. An examination of Cooke's distance and standardized residuals from the final regression model indicated no condition had a Cooke's distance of greater than 0.24 or a standardized residual greater than 2.4, suggesting there were no significant outliers present. Because of the presence of a significant interaction, three linear regression procedures were conducted as post hoc analyses to examine the nature of the interactions. The results of these procedures is shown in Table 1. The significant interaction terms for each comparison indicated each task type had a different slope, suggesting separate processing speed factors for lexical, nonlexical, and inhibitory tasks. The line of best fit for lexical tasks was y = 1.75x-240, R z =0.75, for non-lexical tasks y = 1.39x+7, R2=0.93, and for inhibitory tasks y = 2 . 3 0 x - 5 8 5 , R2=0.88 (see Figs. 1-3). It should be noted, however, that overall the data were well fit with a single linear regression model, y = 1.50x - 52, R 2 = 0.87. Uneven distribution of reaction time values has
44
J. Schatz / Schizophrenia Research 30 (1998) 41 49
Table 1 Hierarchical regression procedures comparing the regression lines for lexical, nonlexical, and inhibitory processing speed equations for schizophrenia Independent variable
df
R2
IR 2
F
A. Lexical vs non-lexical step 1: RT-comparison group step 2: task type step 3: RT x task type
l 177 1 776 1 175
0.88 0.88 0.89
0.88 0.00 0.01
1340.66"* 0.54 12.90"*
B. Lexical vs inhibitory step 1: RT-comparison group step 2: task type step 3: RT x task type
1 102 1 101 I 100
0.80 0.80 0.82
0.80 0.01 0.01
402.36** 2.78 10.11"*
C. Nonlexical vs inhibitory step 1: RT-comparison group step 2: task type step 3: RT x task type
l 107 1 106 1 105
0.91 0.91 0.94
0.91 0.00 0.03
1022.19"* 5.74* 40.89**
*p < 0.05.
**p<0.01. 2500
25007
[I i
2o00J
%
2000-
~ ,5oo!
Isoo~ z /
.~
.~ ,ooo]
I000~
500
~ [] D ,
© 0
!
5OO
j
/ ,/ /,
0
/
0 500
1000
1500
2000
2500
RT (mse¢) - comparison group
0
500
1000
1500
2000
2500
RT (mscc) - comparison group
Fig. 1. Scatterplot of reaction time values showing the lines of best fit for non-lexical cognitive tasks. The linear model of 1.39x+7 accounts for 93% of the variance. The slope differs significantly from 1, t(88)= 12.29, p<0.01, indicating slowing on non-lexical tasks relative to the comparison group.
Fig. 2. Scatterplot of reaction time values showing the lines of best fit for lexical cognitive tasks. The linear model of 1.75x-240 accounts for 75% of the variance. The slope differs significantly from 1, t(85)=6.77, p<0.01, indicating slowing on lexical tasks relative to the comparison group.
b e e n r a i s e d as a p o t e n t i a l c o n f o u n d w h e n c o m p a r ing d i f f e r e n t d o m a i n s o f r e a c t i o n t i m e t a s k s in general slowing models. Because the reaction time values for non-lexical tasks had a number of s h o r t e r r e a c t i o n t i m e v a l u e s t h a n e i t h e r t h e lexical o r i n h i b i t o r y t a s k s (i.e., r e a c t i o n t i m e v a l u e s in the c o m p a r i s o n g r o u p o f less t h a n 350 m s ) , the regres-
s i o n a n a l y s e s w e r e r e p e a t e d after d r o p p i n g t h e s e exceptionally short reaction time conditions from the non-lexical domain. This created equal ranges o f r e a c t i o n t i m e v a l u e s in the c o m p a r i s o n g r o u p s a c r o s s all t h r e e d o m a i n s . T h i s m a t c h i n g p r o c e d u r e , however, had a minimal impact on the hierarchical r e g r e s s i o n p r o c e d u r e s : the i n h i b i t o r y d o m a i n still
J. Schat: / Schizophrenia Research 30 (1998) 41-49 2500
2000-
,2.
©
1000 ~
/
g
/ A /a
, 500
/
/
,/ 500
1000
1500
2000
2500
RT (msec) - comparison group
Fig. 3. Scatterplot of reaction time values showing the lines of best fit for inhibitory cognitive tasks. The linear model of 2.30x-585 accounts for 88% of the variance. The slope differs significantly from 1, t(15)=6.06, p<0.01, indicating slowing on inhibitory tasks relative to the comparison group. differed from the lexical and non-lexical domains; the difference between the lexical and non-lexical domains, however, only reached a trend level ( p < 0 . 0 6 ) . This matching procedure did not significantly alter the processing speed equation for the non-lexical domain (y = 1.36x + 18). A second set of analyses was conducted to explore possible relationships between generalized slowing and particular demographic or clinical aspects of the groups with schizophrenia. From the 198 conditions, the following variables were coded for each group: age of the schizophrenia group (available for 198 conditions); years of education in the schizophrenia group (available for 135 conditions); percent of males in the sample (available for 168 conditions); mean total score on the Brief Psychiatric Rating Scale for the sample (BPRS, available for 77 conditions); the percent of individuals classified as having paranoid features (available for 42 conditions); and the percent of the sample on antipsychotic medication at the time of the study (available for 145 conditions). Standardized residuals for each condition were saved from the linear regression model exploring generalized slowing (i.e., the first regression analysis reported above). These values represent the extent to which each reaction time value was
45
slower (or faster) than predicted from the overall generalized slowing model. Correlation coefficients between these standardized residuals and the demographic and clinical variables indicated relatively slower response times in the schizophrenia group compared with the comparison group were associated with lower education, r = - 0.47, p < 0 . 0 1 , higher BPRS scores, r--0.48, p<0.01, a lower percent of paranoid-type individuals, r = - 0.52, p < 0.01, and a higher percent of individuals on antipsychotic medications, r=0.30, p <0.01. Finally, an analysis was conducted of the abovementioned demographic and clinical factors in relation to the cognitive domains sampled in the study. This analysis evaluated if the domain specific patterns of cognitive slowing could be due to an interaction between characteristics of the patient population and the type of task used in the study. Each demographic or clinical variable was used as the dependent variable and the cognitive domain for each condition (lexical, non-lexical, inhibitory) was treated as the independent variable, One-way analysis of variance procedures did not indicate demographic characteristics were related to the type of task used in the studies (all Fs less than 1).
3.
Discussion
The current meta-analysis suggests several points regarding cognitive processing efficiency in schizophrenia. First, generalized slowing appears to be a significant aspect of information processing in schizophrenia, accounting for 87% of the variance in reaction times across studies. Thus, in terms of effect size, generalized slowing in schizophrenia is a highly significant aspect of performance. This general slowing of cognitive processing speed may have important implications for understanding the sources of cognitive dysfunction in schizophrenia. General processing speed has been shown to be a primary mechanism for improvements in working memory capacity in children (Fry and Hale, 1996; Kail, 1992; Kail and Park, 1994). The importance of processing speed for the development of working memory suggests that
46
J. Schatz / Schizophrenia Research 30 (1998) 41-49
understanding the mechanisms responsible for cognitive slowing in schizophrenia could provide a developmental framework for working memory dysfunction in this disorder. The presence of working memory dysfunction in both individuals with schizophrenia and first-degree relatives (Park et al., 1995; Spitzer, 1993) further suggests that abnormalities in processing speed could be a part of the genotype of the disorder. Further study is needed to evaluate these hypotheses. A second implication of this meta-analysis is that, although a single factor accounts for a large proportion of the variance, the data are better fit by also considering three domains of cognition. Thus, the extent of cognitive slowing appears to be dependent, in part, on the particular type of information processing. The domain specific aspects of slowing may be of theoretical and clinical importance, but appear to be small in terms of effect size. There were no significant outlier conditions present in this meta-analysis, which suggests that although domains may differ in degree of slowing, there was no evidence for a specific cognitive task which showed disproportionate slowing beyond domain specific effects. Certainly, not all potential cognitive domains or specific cognitive processes were represented in the meta-analysis. Nonetheless, the data suggest that not all classes of cognitive processes are impacted evenly in schizophrenia. This is counter to a hypothesis that the cognitive dysfunction in schizophrenia is due to a single, generalized deficit. A third implication of the results of this study is that the pattern of slowing among these three domains has not been observed previously in groups with significant cognitive slowing, such as older adults (Hale et al., 1987; Lima et al., 1991). This suggests the pattern of cognitive slowing in schizophrenia may be a unique configuration of inefficiencies that differs from other types of cognitive slowing. The particularly inefficient processing on tasks which required inhibition, or ignoring competing stimuli, is consistent with previous suggestions that the breakdown of normal selective attention mechanisms is an important aspect of cognition in schizophrenia (Shakow, 1963). There are several points to be made regarding the phenomenon of generalized slowing in schizo-
phrenia. First, as has been done in adult aging, it will be useful to experimentally validate the present pattern of findings. Second, although cerebral dysfunction is associated with cognitive slowing (Ferraro, 1996), little is known about the relationship between domain specific slowing and cerebral dysfunction. Future studies using functional imaging or examining individuals with focal injury may be useful for understanding the neurologic underpinnings of domain specific slowing in schizophrenia. Learning more about these systems could help further illuminate some specific types of cerebral dysfunction in schizophrenia. Finally, because schizophrenia is a heterogeneous disorder, it may be useful to explore variability in the pattern of cognitive slowing across domains in relation to other factors, such as symptomatology or variations in the course of the disorder. An exploratory analysis of such factors in the present meta-analysis suggested there may be aspects of the severity and symptomatology of schizophrenia associated with the extent of generalized slowing. The inconsistent reporting of these variables in the studies reviewed in this report, however, makes it difficult to draw strong conclusions about the precise nature of these relationships. This meta-analysis also suggests several methodological considerations when using reaction time to measure cognitive performance in schizophrenia. As has been discussed with accuracy based measures of cognition (e.g., Chapman and Chapman, 1989), it is extremely important to consider task difficulty, in addition to task content. If only a limited number of reaction time measures are to be used, specific cognitive abilities will be best understood if one matches different cognitive tasks for response time in a healthy comparison group. A second strategy to control for cognitive slowing is to identify cognitive processes whereby poorer performance is associated with faster response times. This strategy for separating out generalized slowing from task specific deficits may prove particularly fruitful when studying normal inhibitory processes, as one may expect faster reaction times with defective inhibition across a number of cognitive paradigms (see Dagenbach and Carr, 1994). It is also recommended that researchers expand the range of cognitive reaction
J. Schatz / Schizophrenia Research 30 (1998) 41 49
time tasks used in individual studies. This would strengthen conclusions in individual studies that any task specific effects found are truly above and beyond what is expected from generalized slowing.
Appendix A Studies included in meta-analysis
Bruder, G., Sutton, S., Babkoff, H., Yozawitz, A., Fleiss, J.L., 1975. Auditory signal detectability and facilitation of reaction time in psychiatric patients and nonpatients. Psychological Medicine, 5, 260-272. Bruder, G., Yozawitz, A., Berenhaus, I., Sutton, S., 1980. Reaction time facilitation in affective psychotic patients. Psychological Medicine, 10, 549-554. Burch, J.W., 1995. Typicality range deficit in schizophrenics' recognition of emotion in faces. Journal of Clinical Psychology, 51, 140-52. Carter, C.S., Robertson, L.C., Nordahl, T.E., 1992. Abnormal processing of irrelevant information in chronic schizophrenia: selective enhancement of Stroop facilitation. Psychiatry Research, 4l, 137-146. Carter, C.S., Robertson, L.C., Nordahl, T.E., O'Shora-Celaya, L.J. and Chaderjian, M.C., 1993. Abnormal processing of irrelevant information in schizophrenia: the role of illness subtype. Psychiatry Research, 48, 17-26. Chapin, K., Wightma, L., Lycaki, H., Josef, N., Rosenbaum, G., 1987. American Journal of Psychiatry, 144, 948-950. David, A.S., 1995. Negative priming (cognitive inhibition) in psychiatric patients: effects of neuroleptics. Journal of Nervous and Mental Disease, 183, 337-339. David, A.S., 1993. Spatial and selective attention in the cerebral hemispheres in depression, mania and schizophrenia. Brain and Cognition, 23, 166-180. Davies-Osterkamp, S., Rist, F., Bangert, A., 1977. Selective attention, breadth of attention, and shifting attention in chronic nonparanoid schizophrenics. Journal of Abnormal Psychology, 86, 461-469.
47
Elkins, I.J., Cromwell, R.L., 1994. Priming effects in schizophrenia: associative interference and facilitation as a function of visual context. Journal of Abnormal Psychology, 103, 791-800. Gray, A. L., 1975. Autonomic correlates of chronic schizophrenia: a reaction time paradigm. Journal of Abnormal Psychology, 84, 189-196. Hermanutz, M., Gestrich, J (1991. Computerassisted attention training in schizophrenia. European Archives of Psychiatry and Clinical Neuroscience, 240, 282-287. Hirt, M., Pithers, W., 1990. Arousal and maintenance of schizophrenic attention. Journal of Clinical Psychology, 46, 15-20. Hirt, M., Pithers, W., 1991. Selective attention and levels of coding in schizophrenia. British Journal of Clinical Psychology, 30, 139-149. John, C.H., Hemsley, D.R., 1992. Gestalt perception in schizophrenia. European Archives of Psychiatry and Clinical Neuroscience, 241, 215-221. Knight, R.G., Youard, P.J., Wooles, I.M., 1985. Visual information-processing deficits in chronic schizophrenic subjects using tasks matched for discriminating power. Journal of Abnormal Psychology, 94, 454-459. Koh, S.D., Szoc, R., Peterson, R.A., 1977. Shortterm memory scanning in schizophrenic young adults. Journal of Abnormal Psychology, 86, 451-460. Kopp, B., Mattler, U., Rist, F., 1994. Selective attention and response competition in schizophrenia patients. Psychiatry Research, 53, 129-139. Koyama, S., Hokama, H., Miyatani, M., Ogura, C., Nageishi, Y., Shimokochi, M., 1994. Electroencephalography and Clinical Neurophysiology, 92, 546-554. Koyama, S., Nageishi, Y., Minoru, S., Hokama, H., Yoshikazu, M., Miyatani, M., Ogura, C., 1991. The N400 component of event-related potentials in schizophrenia patients: a preliminary study. Electroencephalography and Clinical Neurophysiology, 78, 124-132. Laplante, L., Everett, J., Thomas, J., 1992. Inhibition through negative priming with Stroop stimuli in schizophrenia. British Journal of Clinical Psychology, 31,307-326.
48
J. Schatz / Schizophrenia Research 30 (1998) 41-49
Maier, W., Franke, P., Kopp, B., Hardt, J., Hain, Ch., Rist, F., 1994. Reaction time paradigms in subjects at risk for schizophrenia. Schizophrenia Research, 13, 35-43. Min, S.K., Oh, B.H., 1992. Hemispheric asymmetry in visual recognition of words and motor response in schizophrenic and depressive patients. Biological Psychiatry, 31,255-262. Ober, B.A., Vinogradov, S., Shenaut, G.K., 1995. Semantic priming of category relations in schizophrenia. Neuropsychology, 9, 220-228. Pharr, D.R., Connor, J.M., 1977. Similarities and differences in encoding processes in chronic schizophrenics and normals. Perceptual and Motor Skills, 45, 43 t-443. Pharr, D.R., Connor, J.M., 1980. Memory scanning in a visual search task by schizophrenics and normals. Journal of Clinical Psychology, 36, 625-631. Rist, F., Thurm, I., 1984. Effects of intramodal and crossmodal stimulus diversity on the reaction time of chronic schizophrenics. Journal of Abnormal Psychology, 93, 331-338. Rockstroh, B., Muller, M., Wagner, M., Cohen, R., Elbert, T., 1994. Event-related and motor response to probes in a forewarned reaction time task in schizophrenic patients. Schizophrenia Research, 13, 23-34. Rosenbaum, G., Shore, D,, Chapin, K., 1988. Attention deficit in schizophrenia and schizotypy: marker versus symptom variables. Journal of Abnormal Psychology, 97, 41-47. Rosofsky, I., Levin, S., Holzman, P.S., 1982. Psychomotility in the functional psychoses. Journal of Abnormal Psychology, 91, 71-74. Schwartz, F., Carr, A.C., Munich, R.L., Glauber, S., Lesser, B., Murray, J., 1989. Reaction time impairment in schizophrenia and affective illness: the role of attention. Biological Psychiatry, 25, 540-548. Spitzer, M., Braun, U., Maier, S., Hermle, L., Maher, B.A., 1993. Indirect semantic priming in schizophrenic patients. Schizophrenia Research, 12, 71-80. Spitzer, M., Braun, U., Hermle, L., Maier, S., 1993. Associative semantic network dysfunction in thought-disordered schizophrenic patients:
direct evidence from indirect semantic priming. Biological Psychiatry, 34, 864-877. Spitzer, M., Weisker, I., Winter, M., Maier, S., Hermle, L., Maher, B.A., 1994. Semantic and phonological priming in schizophrenia. Journal of Abnormal Psychology, 103, 485-494. Strandburg, R.J., Marsh, J.T., Brown, W.S., Asarnow, R.G., Futhrie, D., Higa, J., 1991. Reduced attention-related negative potentials in schizophrenic children. Electroencephalography and Clinical Neurophysiology, 79, 291 307. Vensky-Stalling, I., Mussgay, L., Cohen, R., 1985. Cognitive dysfunction in chronic schizophrenia on a letter matching task by Posner. Zeitschrift fur Klinische Psychologie, 14, 228 237. Vinogradov, S., Ober, B.A., Shenaut, G.K., 1992. Semantic priming of word pronunciation and lexical decision in schizophrenia. Schizophrenia Research, 8, 171-181. White, C., Farley, J., Charles, P., 1987. Chronic schizophrenic disorder II. Reaction time, social performance and arousal. British Journal of Psychiatry, 150, 374 379. Zahn, T.P., Carpenter, W.T., McGlashan, T.H., 1981. Autonomic nervous system activity in acute schizophrenia: method and comparison with normal controls. Archives of General Psychiatry, 38, 251-258. Zahn, T.P., Rumsey, J.M., Van Kammen, D.P., 1987. Autonomic nervous system activity in autistic, schizophrenic, and normal men: effects of stimulus significance. Journal of Abnormal Psychology, 96, 135-144.
References
Cerella, J., 1985. Information processing rates in the elderly. Psychol. Bull. 98, 67 83. Chapman, L.J., Chapman, J.P., 1989. Strategies for resolving the heterogeneityof schizophrenics and their relatives using cognitive measures. J. Abnorm. Psychol. 98, 357-366. Dagenbach, D., Carr, T.H., 1994. Inhibitory Processes in Attention, Memory, and Language. Academic Press, San Diego. Ferraro, R.F., 1996. Cognitive slowing in closed head injury. Brain and Cognition 32, 429M40. Fry, A.F., Hale, S., 1996. Processingspeed, working memory, and fluid intelligence: evidencefor a developmentalcascade. Psychol. Sci. 7, 237 241.
J. Schatz / Schizophrenia Research 30 (1998) 41-49
Gjerde, P.F., 1983. Attention capacity dysfunction and arousal in schizophrenia. Psychol. Bull. 93, 57-72. Hale, S., 1990. A global developmental trend in cognitive processing speed. Child Development 61,653 663. Hale, S., Myerson, J., 1996. Experimental evidence for differential slowing in the lexical and nonlexical domains. Aging, Neuropsychology, and Cognition 2, 154 165. Hale, S., Myerson, J., Wagstaff, D., 1987. General slowing of nonverbal information processing speed: evidence of a power law. J. Gerontol. 42, 131 136. Kail, R., 1991. Developmental change in speed of processing during childhood and adolescence. Psychol. Bull. 109, 490 501. Kaik R., 1992. Processing speed, speech rate, and memory. Dev. Psychol. 28, 899 904. Kail, R., Park, Y., 1994. Processing time, articulation time, and memory span. J. Exp. Child Psychol. 57, 281 291. Lima, S.D., Hale, S., Myerson, J., 1991. How general is general slowing? Evidence from the lexical domain. Psychol. Aging 6, 416 425.
49
Nebes, R.D., Madden, D.J., 1988. Different patterns of cognitive slowing produced by Alzheimer's disease and normal aging. Psychol. Aging 3, 102-104. Park, S., Holzman, P.S., Goldman-Rakic, P.S., 1995. Spatial working memory deficits in the relatives of schizophrenic patients. Arch. Gen. Psychiatry 52, 821-828. Randolph, C., Goldberg, T.E., Weinberger, D.R., 1993. The neuropsychology of schizophrenia. In: Heilman, K.M., Valenstein, E. (Eds), Clinical Neuropsychology. Oxford University Press, New York, pp. 499-522. Salthouse, T.A., 1985. A Theory of Cognitive Aging. NorthHolland, Amsterdam. Shakow, D., 1963. Psychological deficit in schizophrenia. Behav. Sci. 8, 275-305. Spitzer, M., 1993. The psychopathology, neuropsychology, and neurobiology of associative and working memory deficits in schizophrenia. Eur. Arch. Psychiatry Clin. Neurosci. 243, 57 70. Yates, A.J., 1966. Psychological deficit. Annual Review in Psychology 18, 103 117.